Correlation between knee 3D kinematic parameters during gait and ...

Abstracts / Osteoarthritis and Cartilage 22 (2014) S57–S489

Results: Table 1 presents the correlation coefficients between the radiographic grading sub scores and kinematic parameters. Correlations with the frontal plane kinematics showed to be the strongest, with R values going from 0.641 to 0.724. Furthermore, joint space demonstrated to be strongly correlated with frontal plane kinematics based on the high weight of this x-ray sub-score in the MLR. Results in the transverse and sagittal planes showed negative correlations ranging from -0.440 to -0.497. Results show that FO is the sub-score that is the most strongly correlated (in this case negative correlation) with kinematic parameters in the transverse plane. A residues analysis was also conducted to confirm a random distribution of the residues around zero (figure 1). The absence of error patterns means that the choice of a linear regression fits well the relationship between the kinematic parameters and radiographic sub scores. Conclusions: The frontal plane presented the strongest correlation between the x-ray grading scale and the knee kinematic parameters. This analysis showed that the higher the x-ray sub scores are, meaning a more advanced state of the disease, the higher the varus angle parameters during gait are. This strengthens the link between a dynamic varus alignment and medial TF compartment OA progression. Interestingly, joint space showed a strong influence on the correlations with kinematic parameters in frontal plane. The transverse and sagittal plane both showed a negative correlation, meaning that the higher the x-ray sub scores are, the lower the kinematic parameters will be. In the sagittal plane this translate to a more fixed flexion adaptation strategy to protect the knee and in the transverse plan to a more internally rotated tibia. An offset towards internal rotation was previously linked to faster progression of OA which could also support the link with higher FO score. Further investigations, with a larger and more homogeneous sample group should lead to stronger correlations and should better define the relationships between the knee 3D kinematic parameters and radiographic grading system. This study supports the involvement of dynamic biomechanical factors in the progression of the disease. Results give new insights to develop personalized treatment plan based on kinematic parameters and x-ray OA grading scales to potentially limit the progression of the disease.

S97

Purpose: Patellofemoral (PF) osteoarthritis (OA) is known to considerably limit activities of daily living. To assess the severity and progression of PF OA, skyline view radiographs are typically used. Radiographic grading scales have been developed to quantify the degenerative changes. Functional outcome measures (i.e knee kinematics) are also recognized to be good indicators of knee OA impairments. Few studies have been done in order to evaluate a link between radiographic scores and joint function. The objective of this study is to assess if PF knee OA grading scores are correlated with functional knee kinematics parameters. Methods: Twenty-three knees from 17 patients with a confirmed diagnosis of PF OA underwent a knee functional assessment (KneeKG, Laval, Canada) measuring, in clinic, 3D knee kinematic during treadmill gait. All patients had skyline view x-ray of their knees graded using OAISYS radiographic scheme. OAISYS scale is composed of 4 sub scores (JS (joint space), FO (femoral osteophytes), PE (patella erosion) and SU (subluxation)) and one total score (TS). These sub scores are applied to the most damaged knee compartment: medial or lateral. Knee kinematic parameters were evaluated in all three planes (sagittal, frontal, transverse,) going from punctual angle values at specific times of the gait cycle, ranges of motion (ROM), curve slopes and mean values for specific phases of the GC. Using SPSS Statistics software, Pearson’s Correlation Coefficient was first evaluated between TS and each kinematic parameter to find a preliminary result of the correlations. Secondly, kinematic parameters that showed correlation (p